Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
DOCUMENT RESUME
En 205 571 TM 010 456
AUTHOR Yalow, ElRnnnTITLE Individual Differences in Learning from Verbal and
Figural Mater?als. Technical Report No. 12: AptitudeResearch Protect.
INSTITUTION Stanford Univ., Calif. School of Education.SPONS AGENCY Advanced Research Projects Agency (DOD), Washington,
D.C.: Office of Naval Research, Arlington, Va.Personnel and Training Research Programs Office.
PUB DATE Sep BOCONTRACT N00014-75-C-OB92NOTE 90p.
'EDRS PRICEDFSCRIPTORS
MF01/PC04 Plus Postaae.*Academic Ability: Aptitude Tests: *AptitudeTreatment Interaction: *Educational Strategies: HighSchools: Individual Differences: *InstructionalMaterials: Pretests Posttests: *Remedial Instruction:Social Studies: *Time Factors (Learning)
ABSTRACTOf primary interest in this study was the effect of
general ability on learning. It was hypothesized that students higherin general ability would obtain higher posttest scores on theaveragethan lower ability students, and 'that verbal and figural explanatorysupplements to minimal instructional materials would reduce theregression of general ability on outcome. It was expected thatstudents with higher aptitude scores would obtain higher posttestscores. The effects of a more task-specific aptitude, graphprocessing, were explored, and involved both, immediate and delayedlearning outcome measures. A course in Economics was presented tohigh-school students using one of three sets of instructionalmaterials. Before the course, participants took a three-hour aptitudebattery and were randomly-assigned to treatment. Posttests wereadministered at the end of the course and two weeks later.Generalized regression analysiS was used to assess the effects ofaptitudes, treatments, and interactions. This study provided evidencethat neither aptitude nor instructional treatment alone can fullydescribe learning outcomes. Further, instructional supplements, canbe effective in filling in for student weaknesses and reducingdifferences between high and low ability students. (Author/GK)
**********************************************************************Reproductions supplied by EDRS are the best that can be made
from the original document.**********************************************************************
INDIVIDUAL DIFFERENCES IN LEARNINGFROM VERBAL AND FIGURAL MATERIALS
ELANNA YALOW
TECHNICAL REPORT NO. 12APTITUDE RESEARCH PROJECT
SCHOOL OF EDUCATIONSTANFORD UNIVERSITY
Sponsored by
Personnel and Training Research ProgramsPsychological Sciences Division
Office of Naval Research
and
Advanced Research Projects Agency
under
Contract No. N0001,1-75-C-0882
Approved for public release; distribution unlimited.Reproduction in whole or in part is permitted for
any purpose of the United States Government.
SEPTEMBER 1980
DEPARTNIENT OP EDUCATIONNATIONAL INSTITUTE OF EDUCATION
IDUCATIONAL IILfi0lII10E8 INFODMATIONIEHICI
)1 Flub 1100010010 1001 11000 ruprodul.(1 ai,10.10v0,1 110111 Uhl 110101111 III morileallonfitlijiltatim II
I I Milo, r.hanpu6 Ihivo holm mull, In Improuu
1'1111110 (JI vow ill 00111011d ODdud u1 till 1111/1:11
11111111 till nut nucunn,utlY lotimbont ilhrial NIL111181/1111 (II
"PERMISSION TO REPRODUCE THISMATERIAL HAS BEEN GRANTED DY
Otti it ,ot
041-..g toutAt
TO THE EDUCATIONAL RESOURCESINFORMATION CENTER (ERIC)."
INDIVIDUAL DIFFERENCES IN FEARING
FROM VERBAL AND FIGURAL MATERIALS
ELanna Yalow
TECHNICAL REPORT No. 12
APTITUDE RESEARCH PROJECT
SCHOOL OF EDUCATION
STANFORD UNIVERSITY
Sponsored by
Personnel and Training Research ProgramsPsychological Sciences Division
Office of Naval Research
and
Advanced Research Projects Agency
under
Contract No. N00014-75-C-0882
The views and conclusions contained in this document arethose of the author and should not be interpreted asnecessarily representing the official policies, eitherexpressed or implied, of the Office of Naval Research,the Advanced Research Projects Office, or the U.S. Government.
Approved for public release; distribution unlimited.Reproduction in whole or in part is permitted for anypurpose of the United States Government.
September 1980
UNCIASSITIFD- ,
1kt A1411.1t A111114 U1 111,5 P/0,1; (Ilion tIolot................______..........._ .. ... .. ,........
REPORT DOCUMENTATION1 61' 1i4iTh6ihia.ii
11
4 1111.r(...1.,lifill.1
IlidiVidUal niIIP14.11CC!I In Loain
and FIgntal Materials
1 All I 111111( +)
Fianna Yalow
977r irt' 0141.11N G ONG/kW/ A TION 11 AMC AND Aluu4rSe11001 Of EdllealiOnStanford UniversityStanford, California 94305
II. CONTROLLING OFFICE NAME ANN ADDRESS
Personnel kind Training ResearchPsychological Sciences Division,
'AttF
.2 wiy1 Al 1%`,11111 rill.
ng From Ve1 h,11
up..AD FrummccuirniHi! voin: uomvi.viitlii Fnum
i Hu .l 10 11'.11 i s CAT Milo 14114111!N
.... ..
N I .1.1: o1. NI Poll I A 141: HUM I.t)V1,10--11
Technical. Report, . , . . . .
A o9.11(1/1MING ORG. 111.0.ifil ( NUMItt II
I?11 CoNTII ACT ON GRANT NUM111:11(0
N00014-75-C-0882
10. HNoCifIAM EL EMCN T. P110.1FCT, TASKAREA A WORK UNIT WAM:HS
NR 154-376
ProgramONR, 458
12. REPORT DATESeptember 1980
IL NUMBER OF PAGES44
14. MON I TUNING AGENCY NAME A ADDRESS(If ,Iillernt h011) Controlling Office) IS. SECURITY CLASS. (of this report)
UNCLASSIFIED
IS& DECLASSIFICATION/DOWNGRADINGSCHEDULE
16. DISTRIBUTION STATEMENT (of this Report)
UNLIMITED
17, DISTRIBUTION STATEMENT (of the abatract entered in Block 20, If different from Report)
UNLIMITED.
lb. SUPPLEMENTARY NOTES
This research was jointly sponsored by the Office of Naval Research andthe Defense Advanced Research Projects Agency.
19. KEY WORDS (Continue on revora aide II ncrerriery and Identify by block number)
Aptitudes, aptitude-instructional treatment interaction, cognitiveabilities,learning, verbal vs. figural materials in instruction,elaboration of instruction.
20. ABSTRACT (Continue on reverse aide If necessary and Identify by block number)
The effects of supplementary verbal and figural instructional materialson students of different abilities are not fully understood. Findings inthis area have been inconclusive and inconsistent. Rarely have treatment oraptitude specifications been sufficiently precise to relate outcome to partic-ular instructional components. The present study attempted to improve onsome of these shortcomings.
Of primary interest in this study was the effect of general ability onlearning. It was hypothesized that students higher in general ability would
DD 1 JAN k73 EDITION OF I NOV 65 IS OBSOLETE.
N 0107. LF 014- 6601UNCLASSIFIED
SECURITY CLASSIFICATION OF THIS PAGE (Whoa Dots Entered)
4
ori1'rA,i!Hrtr6
4k4,tIHI I V A111141100' 1110 Alt 11441., 11.14 roiatot41
eht 0111 1110101 pita( I 1,1 II 1.111 I :11/1't-,1)%1 Iil,ilt 1,114c ;ill ,(11thqlt ti, It
ow; IIII 11101 11%11111 111.'i 1:',111 111.11 1'0111,11 ,1111 1011 ,11 4)(111:1,1,11111 111,11,,111C111:i 111
111i 111111A 1 In';1Iu4 t fou,ll 111,111,I .11 would I cHhICI' I Ili* 1 'fl14':+14,11 4,1 I,1'11!1 rl 4111 1 11
1111 111111'111111, That I , the Ie lie oceil I ('1 .111,1 h 101 :Willi y t 1111110111 h
who'll ird o eitii !mutt 111 y
verbal et I trill ill d 1'.111,1V
1111:1 !ii 1111k,' 1111 1111'1 1111 4'11 111' 1111 1 1'1 1'111 1.11 111111Th 1 III t' i 411 1 1 '1)111111
;11/1 Y 11;1' ) 11111 1 111111 p.lt 1.11 V1'111.11 1:.11 11111 .11, 1 1 11 Asl 1)1,1 alro II
14,1.4 0:4.110I 111,11 HI ildi'111 h !,he apt 1111,11' 14011111 li111,1ill 1110101
ino:lto!:1 rutihointoro, i f wiei hypetheHi thAt t:c and I;lv might
moderato the relAllow; between instiihilon And outieme different ly.
Ih,ll emeil 1,eci i I too.1 111 1 11 at tntont 1 llrld
I It I:c; .,11 to !own I offiont wo o;tpoct 1,1 I o part It 0,1 11':o n oilool ii
ow Intko explored the ellecui of a more f;ed:-.spiiellie aptitude, wrap',
precelng, en outcome, And Involved both immedi ito 11111 delayed learning out
come 1111',1!:11 I'1':;
A i.otirse 111 Fconowits wa!; piesented to high !whool students using
line oe sets III i 11:;1 luI ;mid I mat er ta lie fore t he course part tolpont::
took o 1-hoor .Aptilnko holtorV and wro tondomly 4o;ot,t;ned to troatmOnt. POnt-
r,,t, wore administered ;It the end of the course and two weeks later.
The Inpitc trootwont covered the theory of MaYkot prio,O, Treatments varied
in the ixplAnAtory di,iplays 'and the difficulty of the processing demands, duo
treAtment (M1N)presented the information with little redundancy, feo examples,
and limited explanations. Another (VF) covered the same material 1 MIN, cd.th
additional verbal expansion materia i. A third (EN) covered the material of
MIN, with additional graphs and diagrams as figural expansion.Ceneralied regression analysis was used to assess the effects of aptitudes,
treatments, and their interitions (ATI). On the immediate posttest, students
in VF and FE did better than Htudents in MIN, suggesting that the elaboration
provided in these conditions helped students learn. Significant ATIs suggest-
ed that the elaboration was particularly useful to low ability students. High
ability students did as well or better in MIN. Thus, the regression of
achievement on general ability was steepest in MIN and reduced in VE and FE.
Partitioning the total test score by posttest item type indicated that VE
was particularly helpful on verbal items and FE was particularly helpful on
figural items. Again, significant ATI indicated that these treatments were
particularly helpful to loan ability students. Hence, the regression of ver-
bal items on general ability was least steep in VE; the regression of figural
items on general ability was least steep in FE.
Examination of learning outcomes on retention, however, led to strikingly
different conclusions. While students In MIN were worse on average achieve-
ment on the immediate posttest, they performed the best on retnetion. Losses
iron) Immediate to delayed posttest were greatest when the 'assistance was most
direct. That is, losses on verbal items were greatest in VE; losses on fig-
ural items were greatest in FF. No significant main effects or ATI were
associated with the differential between Cc and Cfv on either posttest.
in summary, this study provided evidence that neither aptitude nor instruc-
tional treatment alone can fully describe learning outcomes. Interactions
between them exist and were demonstrated. Further, instructional supplements,
whether verbot or figural, can be effective in filling-in for student weak-
nesses and reducing differences between high and low ability students.
Such supplements, however, lint he used with caution. Reducing the diffi-
culty of instructional materials may, indeed, enhance immediate learning, but
these advantages mav short-lived.
S N 0102- 41. 01,1.660)UNCLASSIFIED
5CURITY CLASSIFICATION OF THIS PAGE(Irhn beit Enftored)
PltriAil
Tho Inve:ltlp,Atton 101101.14.d heroin ptil ail on onv.olt
ronmarch rolect almod at tuolortandfur lite nature and
importance of Individual diffecoocen in apflundo for
learning. Requent!i tor infotmation Ink proloct
and for copion of thin or cohou technical icportn
be addre.lann to:
Profeanor Richard E. Snow, Principal invontigator
Aptitude itencarch Project.
School of Education
Stanford University
Stanford, California 94105
ii
4 II' CI )11.1111
vurVAcr I I
ki!iT (IC 17M11'.;
I.L;T
C11:1111 oi
I INTROPUCTION
hAcVgromid
1114! Pro:i4,11(
I I ME.11101) 8
:;omp10
TrolilmotWi
Material,
Aptitude Menfutros
fu:itriictiouat Materials 1.1
Outcome Measure: 12
P-ocedure
tf QF:1111,Th 15
Descriptive Statistic:: 1.5
Aptitude Measures .15
Outcome Measureq 17
Correlations Between Aptitude and Outcome 19
:Tression Analyses on Outcome 21
T3tal Score Ana_jses 22
Part Score Analyses 24
Verbal Items 24
Figural Items 27
Problems 31
Summary of ReE,rerion Analyses 31
iii
Arm t
h 1'1 4.111 (tut . 11
I hi: Y'tc:naui tic:aunft 11 1/
I hit 1.1 I 04 I :t :11 Al. 1111 y
lho 1)1 I ot cut 1...11 1' 11 ", l 11 )11141
1.1141 I,1 I ut l uI CltA141
Ciltit. I IL)) Iowa mid Imp I 11111i 1')
01.11.:11 CNC 1.:'
Alliwnd I It
A. 1,1).)gtoi !;1.milm (I 11)w I))1 lulu), rut() I.. I )11 ooti III lwtt'nApl I I itilori
It. Pout Leta Col :e I al lotui 'JO
C. Curl 4. I At, tout 114.1 wet.)). AO I t 11de ;111(I Ow cum), .1))
it 0
iv
I I , I 1II I .1 ti I I
.41111111.11 ' 1,1 I I 4.1,1 'pc.. I
, I i lu l, . I
=111.1:11.1 lit:u 1:41 I t1111:14 -11111 1411 1 01:11 I iP I III I
1!.:1 I ill V1 1I .11 :,11111i I
ticAtiti :i1141.11 tl Ilr.0 Iitl I gilti Iii C.:11 1;1
CII t u I:11 I PIM 11011.40C111 :101411 I :kill 1111110 :1111I 1.1111 0;11 V:11 ;III I
h. !'ill111111.11 V ttlIW I :it! tl1 ht ttLi:i 11111, 411 11111 :1 I I v-11
/. tutuurirY sit rd cpw I tio 1111 ()I V1'II1 :11 I t1111
. iluuiurll y 4)1 opw I tio Itomrt,Iti ott (d. I I I I
i. ,,itinnukry 4)1 (..pw (tie Itogrytitt Ion 4)1 Olt I t.-111'1 14
V
1 I ...111
1 I I I, 11 111 1111 .11 1,71 1 4-.11 11..3 1
1111.1.41 1111i ..I 1 1,A 1131 11 :.1111 111 ..1111
11(.1.11 11111 111 "l-1 11.11 11 ,.11111 111 1.11A1111
11(11 .11 11111 11! 1 1 1.z) i .1 I 1 1 ctil z) 1
I 11 I 1!111 1 l I (.111:1 1 17 1.It:11'll
Ii Hcl 1.11 i'141I1 C:IIIZA I II :
vi
1 th, i.1o1 1 - It ., 1,11 1.1, I I 1111-, .11.
1114.. III Ic .11 1,1I i.. I tI.3 tt. I 1.1I. I Cr: 1.; III. I .7.,-1
I 1.111' I I 1.'1,1,1. 1-o I I,nlLa.II .111.1 .4I
I 1 III .11 III I -II ,:1 I II-11 I I-1-11 It'I I
.1111 1,1.1111111. ...II I t it I', i t 11. 1,11.3 t 1. 11
110.111110111 110 crl :It I I.1tIri I AII Ic.rts 01t I a I . ALI ftnaInp,,, bac,
III I I V.1411 I .11,15 I II It: I II ('-t 1.1 I III
14,11.11 pi .$4 I it tt i')//). 11"1" 1 hi" " I,1 4 1100.1i 1111 1 4.:%(1,11 I 11.11
can pt ItI 110 at doopot undo t Al And I ng of Apt II lt.It, I11a1 tticI Ilnial 1 1 v,11
moll' , ;hid 1 ho tut o1:101 Ittntt.
nAoltnround
Tho onopI oI goholAl !!lout .11 JhIIItV co1111.:11 Io most current
modols tit intelligence and ability organt.,,ation, and has boon rgrdd
.1 I the most important aptitude for learning from instruction. Mossuros
oi general ability also seem to show the strongest mid most consistent
Al I.. This is, then, the logical place for now AI research to start.
[lie bulk of prior evidence is consistent with the following hypo-
thesis: When instruction places heavy information processing burdens on
learners, the regression of learning outcomes onto general ability differ-
ences is relatively steep; able students do well and less able students
do poorly. In contrast, when an instructional treatment is designed to
relieve some of the information processing burdens on learners by simpli-
fying, structuring, or elaborating the learning task, the regression of
outcome on general ability is relatively shallow; less able students
1
.4i 1 1
s 114. 11.1.1 64 i 1,i
4 11..11, , 1 1 1 1 1 1 i 11,1 1 1
1411.1 :111.1 I. 1 ill 1:1 I Lit:A, I I Ai I I I, I', I { .111-1 1 I.
,11114114,1 1111 4111 :WI 11 :11 ,',114:11 1 ' 1 t .4 1 1 1 .C.1 .11,1 1 1 i' 11. U, f1.111
.111:8 I VI I. 41111 ,! 1 :) t111 1 1 1,,t1 .11,1 1 11 ', <t< ,
:1I111. :111.1 , I Itiit 1141 1111 1114'1 11'11 1111 1 11111 114'114144441 I 1111.1 :11i 1 1 11 '.'
111111 (111 III l i4I 1114)11,i 1 I .31 1,111 11 1:; ,11 I ou ,1 I I 1 I 111 1 t to 111<11 c i ,,11111,111,
',now. III pron./. In Ifni plvtiouf A1:4(Arstlitm. wicA
4'1/M1111141d Ilnld AnAlvtic vfinAff:Affon dimon-ifon.
Aocond, difforooflAI, hvpotho IhAf thc
modtum of InAfincflon (vothA) v'1. ot ApAfIAI) 1.1 Inc
relative strength And woaknwoi of 61v And t;,. In loAlno411 141111111.1 lit' hono
ticIAI, and Isolated at based on this hypothesis have lopotIod ATI.
lint two different and opposing, 1cind:1 ot marching are possible, and both
have been found; matching to the lolunc.r'q !iirength mkbt be honvlicial
in some ways, but matching to the learner's weakness might have compensa-
tory benefits. In one typical study, for example, Peterson and Hancock
(1974) taught students the mathematics of network tracing using either
verbal, symbolic, or figural materials. Aptitude measures were selected
from Guilford's (1967) system to represent these three content areas.
Posttests were administered immediately after instruction, and again
after one and five weeks. The regression on verbal ability was shallower
In the verbal treatment than in the figural or symbolic treatments al:
all three testings, suggesting that low verbal students were compensated
by a verbal treatment. The regressions of outcome on figural and symbolic
ability were shallower when aptitude matched the instructional condition
only on the immediate posttest. On both retention measures, regressions
were steepest for figural and symbolic ability in the figural and symbolic
treatments, suggesting that one should match to strengths.
The results of other studies investigating relations between verbal
and spatial ability in instruction have also been inconsistent. Allison
(1960) provided instruction on concept attainment tasks using either
verbal stimuli and semantic solution rules or geometric stimuli and
classification solution rules. Verbal and spatial aptitude measures
were used. Those higher in verbal ability did better with verbal con-
tent; there was no effect for spatial ability. Bracht (1970) taught
addition of signed numbers using figural or verbal programmed texts.
Numerical, verbal and spatial aptitudes were measured. There were no
significant ATI. Markle (1969) taught crystallography using programmed
texts composed either entirely of words or emphasizing diagrams. The
pattern of correlations of outcome with verbal and spatial measures was
similar in both treatments. In a series of studies by Carry (1967),
Webb (1971), and Eastmaa (1972), students were taught quadratic inequali-
ties using materials designed to capitalize on spatial-visualization.
After reviewing this series, Cronbach and Snow (1977) concluded, "The
three studies together provide only negative evidence on the possible
relevance of vusualization to a piesentation that uses graphs",(p.285).
3
On the tasis of these and other mixed results, Cronbach and Snow
(1977) rejected "the conclusion that spatial treatments demand spatial
ability and that differentiated Guilford abilities will interact with
treatments of the same name" (p. 293). They did not rule out the possi-
bility of positive findings in the future, however, given more powerful
and penetrating analyses.
Gustafsson (1974; 1976) also obtained conflicting results in a
series of studies exploring the verbal vs. figural contrast. Two of
these studies used a text on polar lights as a verbal treatment; the
pictorial, treatment was a slightly reduced text supplemented with illus-
trations. Aptitude measures included a vocabulary test, a reasoning
test, and a spatial-visualization test. Learning was measured by a short
answer test and an essay test. Results on the essay test were not con-
sistent across the two studies. On the short-answer tests in both
studies, the slope of the regression on verbal ability was steeper in
the verbal treatment than in the pictorial treatment; students low in
Gc did best with pictures, especially if they were also high on Gv.
Students in the third study were taught about the heart and the blood
circulation system using either illustrated or unillustrated materials.
Immediate and delayed outcome measures included items assessing verbal,
pictorial, and spatial criteria. Aptitude measures represented Gc and
Gv. There were no substantial ATI for the verbal or spatial criteria.
Although the pictorial treatment was best for everyone, an ordinal inter-
action indicated that this treatment was least advantageous for students
high on Gc and low on Gv.
More-recent research has not changed the picture appreciably. James
and Knief (1978), for example, taught students to determine tie number of
4
I
subsets in a set of elements using a treatment designeu to capitalize
on either Gf or Cc. A sum of Cc and Gf scores represented general
ability, and the difference between the two scores represented the dif-
ferential hypothesis. A pretest and posttest wee administered. There
were significant main effects and ATI with the sum, but neither the
difference score nor any of its interactions were significant. Although
high ability students, on average, outperformed low ability students,
the treatment designed to capitalize on Gc reduced their advantage.
The consistent results for the general-ability hypothesis, and the
inconsistent results for the differential ability hypothesis, are both
understandable in hindsight. The wide variety of instructional treat-
ment contrasts that yield ATI with general ability can be summarized in
terms of variation in amount of information processing demand, but this
is only a crude summary, at best. The demand characteristics of differ-
ent kinds of instruction are not understood in detail, nor is a process
theory of ability for analyzing task demands in relation to individual
differences available. Treatments are usually poorly specified, and
this hampers our pursuit of both the general and the differential-ability
hypothesis. An enormous range of instructional materials have been
labeled "spatial" or "verbal" with little thought about their processing
demands. The presence of figures or pictures does not indicate that a
treatment requires spatial ability. Diagrams can tax ability but they
can also compensate for weakness. Similarly, it is insufficient to
attach global labels to categories of ability. A "spatial" ability test
does not necessarily measure spatial ability (Lohman, 1979a; 1979b).
Aptitude measures should be understood in terms of amount and kind of
processing demand.
There is not likely to be a simple match of aptitudes and treatments
(Cronbach and Snow, 1977; Salomon, 1972). Some kinds of instruction
build upon the learner's capacities or preferences, requiring students
to bring possessed abilities to bear in learning. Alternatively, in-
structional materials may do for learners what they cannot do for them-
selves, and so may reduce ability-outcome correlations. Less able
students might profit from such assistance, whereas able students might
be turned away by it. Further, learners may substitute abilities they
possess for those they lack. Thus, graphic problems might be solved by
either verbal processing strategies or by direct manipulation of lines
and curves.
Thus, the inconsistency and complexity of earlier ATI results seem
due, in part, to the failure to specify requisite abilities for carefully
delineated treatments, or to provide a common process ription for
aptitude and learning tasks. There has also been inadequate considera-
tion given to the multiple ways in which, aptitudes and treatments might
be matched. The notion that students of high spatial ability necessarily-
do better in spatial treatments ignores the complexities of both ability
and instructional material.
The Present Study
The primary hypothesis investigated here related general ability to
learning. First, students higher in general ability were expected to
obtain higher posttest scores on the average than lower ability studOts.s,
Further, both verbal and figural supplements were expected to reduce the .
slope of the regression of outcome on general ability. The effect/of
verbal supplements was expected to be greatest on verbal outcome measures;
the effect of figural supplements was expected to be greatest on figural
outcome measures.6
1 1-.!-.0
The present study also explored the differential impact of Gc and
Gfv on learning. It was hypothesized that students high in either Gc
or Gfv would learn more than lower ability students. It was further
hypothesized that Gc and Gfv would moderate the relations between instruc-
tion and outcome differently. Verbal supplements were expected to be
particularly useful to students low in Gc; figural supplements were
expected to be particularly useful to students low in Gfv.
Finally, the study was planned to examine long-term as well as
immediate learning. A reduction in average scores from immediate to
delayed posttest was expected. The greatest drop in performance was
expected when instructional supplements were used. By reducing process-
ing demands, supplements might enhance short-term learning while reducing
long-term learning. This effect would be particularly evident where
instructional content and outcome were matched. That is, losses on
verbal outcome measures would be greater when verbal supplements were
used than when figural supplements were used; losses on figural outcome
measures would be greater with figural supplements.
To summarize, this study assessed the relations among aptitudes,
instructional supplements, and learning outcomes. Gc and Gfv were the
aptitudes of particular interest, although more specific aptitudes were
also included. Instructional materials differed in the use of verbal
and figural supplements. Outcome measures distinguished verbal from
figural responses. In general, this study was intended to illuminate
the relations between aptitude and instructional treatment.
7
.A."1
CHAPTER 2
METHOD
A 2-week course in Economics was presented to high school students
using one of three sets of instructional materials. Beforehand, partici-
pants completed a 3-hour aptitude test battery and were randomly assigned
to treatment. One posttest was administered at the end of the course;
another was given two weeks later.
Sample
Participants were recruited from three Palo Alto, California, high
schools. Tenth- and eleventh-grade students responded to an advertise-
ment in\a kcal newspaper and were paid an hourly fee for their partici-
pation. The initial 3-hour aptitude session included a 10-item screening
test to eliminate those already familiar with the instructional content.
Of the 146 students who initially responded, 132 were retained and
completed the experiment. The final sample included 86 females and 46
males; 44 participants were assigned to each of the three conditions.
Treatments
The basic instruction covered the theory of supply and demand,
determination of market price, elasticity of supply and demand, and the
application of these principles to price floors and ceilings, taxation,
and agricultural problems. Materials were adapted from introductory
college economics textbooks (Lipsey and Steiner, 1969; Samuelson, 1976;
Spencer, 1977; Sutton, 1976), but presented at a level appropriate for
high school.
The same material was covered in each treatment condition. Treat-
ments varied, however, in the explanatory displays and the difficulty of
the processing demands. The three instructional conditions were Minimal
8
1 n
(MIN), Verbal Elaboration (VE), and Figural Elaboration (FE).
Information in MIN was presented with little redundancy, few
examples, and limited explanations. Participants were encouraged to
solve problems on their own and to generate their own explanations for
facts and principles. Principles were presented with limited verbal
explanations and figural displays.
VE covered the same basic information as MIN, but with additional
verbal material. Examples were given, verbal explanations were presented,
and basic concepts were redefined as learners encountered new material.
Figural content was identical to that of MIN.
FE also covered the same basic MIN material, but with additional
graphs and diagrams. Examples and exercises using graphs in problem
solving were added. Additional verbiage was used only to help students
understand and manipulate diagrams. The differences among treatments are
summarized in Table 1.
Each treatment consisted of eight 50-minute instructional sessions.
Participants were limited to one instructional session per day.
Materials
Aptitude Measures
Four tests were selected tgmeasure\Gfv: The Advanced Progressive
Matrices Test (Raven, 1962), Paper Foldin Test (French; Ekstrom, and
Price, 1963), Copying Test (French et al., 1963), and Memory for Designs
'(Graham and Kendall, 1960). Measures of Gc included the Terman Concept
Mastery Test. (Terman, 1956), Advanced Vocabulary Test V-4 (French et al.,
1963),.and a fill-in vocabulary test adapted from the Wechsler AdultN
Intelligence Scale (Wechsler, 1955). The latter consisted of 20 words
9
TABLE 1
Summary of Treatment Specifications
Treatment
Component of Instruction MIN VE
General information
Basic,statement of economic principles
Verbal
Figural
Explanation of principles
Verbal
Figural
Examples
Practice problems
Solutions,tu practice problems
Explanations of solutions for practice problems
Verbal
Figural
Redundancy
Verbal
Figural\
Underlining
N
Note. "+" indicates -a component present in the treatment
"-" indicates a component not present or used minimally
10
from the WAIS vocabulary sectior, representing the full rang of item
difficulty. Items were scored using the guidelines presented in the
test manual.
A test of graph processing (GRAPH) was designed and administered to
supplement the broader ability tests. This test measured the ability to
read graphs and to interpret data presented figurally. Items required
either translating verbal information to graphs or giving verbal des-
criptions to interpret graphs.
Another instrument designed specifically for this study was the
Cognitive Preference Questionnaire. This questionnaire asked students
if they preferred learning from verbal material or by reasoning about
diagrams and figures. Attitudes toward selected instructional features
and learning strategies were also solicited.
Instructional Materials'
Three workbooks, corresponding to MIN, VE, and FE, were developed.
Each workbook was composed of eight 10-20 page packets and introduced
approximately three new topf.cs. Students worked through the workbook
in a prescribed manner, answering questions and solving problems in the
packet.
Each packet began with a Summary Sheet listing the major topics
covered in previous sections. In VE and FE, major points were summarized
in the appropriate mode. Participants in MIN were cued to generate the
summary for themselves.,
The last pages in each packet contained problems relevant to the
material covered during the instructional session. These Problem Sets
were included'to encourage students who completed the material before
11
the end of the session to review it, thereby equalizing students' working
times.
Outcome Measures
An immediate and a delayed posttest were administered after the comple-
tion of each instructional unit. The two posttests were similar in format
and content. Each test consisted of 40 items and covered most of the material
presented during the instructional period. On each test, 15 items re-
quired students to answer verbal questions, 15 required students to deal
with figural information, and 10 questions required both verbal and
figural explanations. For these 10 items, the student was asked to indi-
cate the explanation given first. Within each of these categories,' items
required either the application of principles to solve a problem, or
simple recall or recognition of information specifically discussed during
instruction. Response formats included multiple-choice, fill in, and
short problems. Approximately one hour was alloted to complete each
posttest. Table 2 lists the parts of each posttest, including item type
(figural vs. verbal), response format, and the number of items.1
Procedure
Instruction began approximately one week after aptitude testing.
Participants attended one 50 minute session for each of four days during
the first week of instruction, and for each of five days during the
second week of instruction. Students completed one packet of material
during each session. Sessions were held hourly between 3 p.m. and 11 p.m.
on weekdays, and between 9 a.m. and 5 p.m. during the weekend. Students
12I' .e)titi
Table 2
Summary of Outcome Measures
ouLcome Measure Description Maximum # of Points
Total Posttest
Verbal Total
Part 1Part 2
Figural Total
Part 1Part 2Part 3
ProblemsVerbal correctFigural correctVerbal firstaFigural firsta
Total number of correct items
Total number of correct verbal items
Fill-in verbal itemsMultiple-choice verbal items
Total number of correct figural items
Fill-in figural items ; draw figure
Fill-in figural items; interpret figureMultiple-choice figural items
Total number of correct problemsNumber of correct verbal explanations on problemsNumber of correct figural explanations on problemsNumber of correct verbal explanations given firstNumber of correct figural explanations given first
40
15
10
5
15
5
5
5
10
10
10
10
10
Note. Format of Immediate and delayed posttest was identical
Numbers of verbal first and figural first apply only to correct explanations on the problems.
were allowed to schedule themselves freely except that no more than 30
students could be accomodated in a single session.
Upon arrival, students would take a folder bearing their name, and
remove the appropriate packet. At the end of the session, students
returned the packet to the investigator, who checked that the student
had worked on the proper materials. It was not necessary to schedule
participants in the same treatment for the same hours as all students
worked individually.
The immediate posttest was administered to each student after all
instructional materials were completed. All participants took the immediate
posttest on the Friday of the second week of instruction. Most partici-
pants returned for the delayed posttest two weeks later; those who could
not were scheduled individually for their delayed posttest. All partici-
pants completed the delayed posttest within 11 to 17 days after the
immediate posttest.
r-
14
0 k-1
CHAPTER 3
RESULTS
Descriptive Statistics
Aptitude Measures
Means, standard deviations, reliabilities, and courelaticns dmon
aptitude measures for the entire sample are presented in Table
Similar tables for the separate samples in MIN, VE, and FE are presented
in Appendix A.
Scores from the two vocabulary tests and the Terman Concept Mastery
Test were standardized in the total sample and combined to form a com-
posite labeled Gc. The Gfv composite included standardized scores for
Copying, Memory for Designs, Paper Folding, and the Advanced Progressive
Matrices Test. Although GRAPH showed high correlations with both com-
posites it was left as a separate third aptitude since it was thought to
be specifically relevant to this instructional setting. About 42% Of
its variance was estimated to be specific.
Tests included in the Gc composite showed higher correlations with
each other than with measures of Gfv. Copying and Memory for Designs
had higher correlations with other measures of Gfv than with indicators
of Gc. As expected, however, Paper Folding and the Advanced Progressive
Matrices Test, both complex measures of Gfv, showed higher correlations
with Gc than did Memory for Designs and Copying.
The Gc and Gfv composites were combined to form two orthogonal
indices to investigate their combined and differential importance: SUM,
the sum of the Gc and Gfv composites represented general ability, and
DIFF, Gc minus Gfv represented the ability profile difference. Positive
values on DIFF thus indicate students higher in Gc than Gfv, and negative
15
Table 3
Means, Standard Deviations, and Correlaticns ofAptitude Measures for Total Sample (N - 146)
Variable Mean S.D. (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
b(1) Vocabulary Multiple Choice 15.11 4.58 80 68 67 26 11 33 38 36 88 36 71 55
(2) Vocabulary Fill In 26.62 5.76 78a
71 30 27 49 54 53 90 52 81 39
(3) Terman Concept Mastery 30.10 5.10 78a
36 32 49 54 56 89 56 83 34
(4) Copying 32.06 9.41 88b 37 49 53 40 35 79 65 -46
(5) Memory for Designs 17.00 2.09 60a
31 37 38 26 68 53 -43
(6) Paper Folding 13.45 3.50 80b
53 50 49 77 71 -29
(7) Advanced Progressive Matrices 23.64 5.47 83a 62 54 80 77 -27
(8) GRAPH 22'.64 5.50 86a 54 62 66 -08
(9) Gc .00 1.00 91 54 88 48
(10) Cf./ .00 1.00 93c
88 -48
(11) SUM .00 1.76 95c 00
(12) DIFF .00 .96 83c
Note. Decimals omitted from correlations.Reliabilities appear in the main diagonal.
aReliability estimate coefficient a.
bSplit-half reliability estimate.
cReliability estimated as composite.
values indicate an advantage in Gfv. This procedure has advantages over
using Cc and Gfv directly in the analysis (Cronbach and Snow, 1977). First,
hypotheses can he ordered so that the general-ability hypothesis can be
tested independently from the more exploratory differential-ability
hypothesis. Second, SUM and DIFF are uncorrelated, whereas, Gc and Gfv
usually correlate (r 3 .54 in this study). Thus, using SUM and DIFF
prov._asless ambiguous interpretations than using Gc and Gfv directly.
Although students were randomly assigned, the equivalence of groups
was checked. There were no significant mean or variance differences
among the treatment conditions on any of the aptitude measures and only
minor differences in the pattern of correlations in the three treatments.
Thus, no systematic aptitude differences among the three groups could be
identified.
Outcome Measures
Reliabilities of the immediate and delayed posttest were estimated
at .82 and .84, respectively, using coefficient alpha. Correlations
between total scores and major part scores on the immediate and delayed
posttests are presented in Appendix B.
Means and standard deviations for all parts of both posttests are
reported in Table 4. While treatment differences on the immediate
posttest were not always large, group averages in VE and FE were consist-
ently higher on items corresponding to the type of assistance the group
had received. On the average, students in VE and FE also obtained higher
scores on the problems than students in MIN. Thus, the mean total post-
test score was higher for VE and FE than for MIN.
This pattern was not found in the delayed posttest means, also
reported in Table 4. Here, MIN showed a slight overall advantage over
17
r).1/4.0
Table 4
Meann and Standard Deviations on PontAent
Immediate PoulAest. Delayed Pouttest
Treatment MIN V'. FE MIN VE FE
Outcome Mean S.D. Mean S.D. Mean S.D. Mean S.D. Mean S.D. Mean S.D.
Total Posttest 26.27 7.20 29.52 5.25 28.34 4.66 28.89 6.80 27.98 6.18 26.39 6.39
Verbal Total 10.11 2.94 11.89 1.82 10.23 2.11 10.91 2.30 10.09 1.96 10.11 2:22Part 1 7.43 1.68 8.48 1.30 7.41 1.57 7.57 1.93 6.95 1.61 7.16 1.79Part 2 2.68 1.58 3.41 1.00 2.82 1.11 3.34 .96 3.14 .85 2.95 1.03
Figural Total ,.57 2.71 10.50 2.62 11.20 1.71 11.11 3.08 10.82 2.74 9.48 2.81Part 1 3.52 1.27 4.09 1.01 4.27 .85 3.93 1.28 3.84 1.01 3.57 1.15Part 2 _ 3.95 1.06 4.05 .91 4.09 .80 3.98 1.15 3.57 1.00 3.05 .99Part 3 2.09 1.14 2.36 1.12 2.84 1.03 3.20 1.27 3.41 1.26 2.86 1.29
Problems 6.59 2.67 7.14 2.11 6.91 2.17 6.86 2.46 7.07 2.29 6.80 2.54) Verbal Correct 4.02 3.24 5.52 2.83 4.09 3.23 4.61 3.24 5.34 3.28 5.14 3.32
Figural Correct 3.55 2.97 3.43 2.67 3.64 2.74 4.25 3.37 4.91 3.45 4.95 3.54Verbal First 2.14 2.46 4.20 2.83 2.52 2.52 2.25 2.66 2.70 2.75 2.59 2.74Figural First 1.80 2.48 .98 1.68 1.95 2.06 2.50 3.02 2.48 2.54 2.98 3.14
VE and FE on total posttest :irate. Thus, although elaborating and
simplifying instruction (as In VE dIld FE) apparently ptoduced immediate
galns, information acquired through such Instruct: Lou was nut: retained as
well as Lt was In MEN. MIN actually showed a small average gain lrom
immediate to delayed tests, while VE and FE showed losses. The greatest
Losses occurred when treatment and posttest item type were alike. That
is, the greatest losses on verbal items appeared in VE; the greatest
losses on figural items appeared in FE.
On the immediate posttest, there were small differences among groups
in the type of explanation given on the problems. On average, students
in VE gave correct verbal explanations more frequently than did students
in either MIN or FE. Verbal explanations were also more likely to be
given before figural explanations by students in VE. Similarly, in FE,
correct figural explanations to the problems were, on average, given
before verbal explanations. Similar trends were found on the delayed
posttest.
Correlations Between Aptitude and Outcome
Table 5 gives selected correlations between outcome and aptitude
measures. (See Appendix C for complete correlation matrices.)
Significant correlations between SUM and all outcome measures
appeared in all treatments, while correlations for DIFF never differed
from zero. The correlations between total posttests and SUM were similar
in the three treatments, but part scores on the immediate posttest
showed some variation across treatments, particularly when item type and
treatment matched. So, for example, SUM and the immediate verbal total
correlated .50 in VE, but .74 and .72 in MIN and FE, respectively.
Similarly, SUM and the immediate figural total correlated .50 in FE, but
19
0 0
Table ')
Corralatlutm lletweett Selected Aptitude ant Putteai VaViAlvu
'!'reatmout
--- AptitudeOutcome Measure----:.
SUM
MI N
UHF GRAPH SUM 1)1. Fl GRAPH SUM
FE
DLFF
Immediate Posttest
Total Score 83* 04 69* 75* 02 64* 79* 12 65*Vethal Total 74* -05 63* 50* 00 38* 72* 05 65*Figural Total 68* 04 65* 69* -02 64* 50* 22 39*Problems 73* 13 57* 59* 06 47* 60* 03 46*
Delayed Posttest
Total Score 78* 03 71* 83* 00 62* 76* 03 70*Verbal Total 54* 05 38* 72* 01 42* 52* 04 58*Figural Total 70* -04 76* 78* -07 61* 69* 08 54*Problems 77* 09 65* 69* 07 58* 70* -06 66*
Note. Decimals omitted from correlations
*2 less than .05
f111 :UI' f1'1 111 NITA Altd Vt.. .0
colavb!IIIina wIlh 11t1 nolo vIiInnIIV IdonthaI In V. and Vt. antiowhat
!MIA I I 4! I I 11. hi I Ito do III votI lemil 1 0;$1 , I Ili' I III 1 0 I .t1 I ton Ill! WOVII
'AN and iho 1'1 11:11 WAri Iar1',0:1( III vr.
Cnirolallnun holwoon OkAPH iutit tint Imilioafalo unIctuo wcio gonolally
lower 111,in those holweon ;1h4 and onloomo. Thoro wore, howover, idle)
OWtt r t orre I .11 mill hot wool, (;:tAni 1111(1 I initivd I It( 1. (' ()I1W WIWI% I VIII.. alit(
treatment wore mathod, On the delayed ponttent, there wore nnIy
differences among the treatments In the correlations of GRAPH with total
posttest score and the probiems. The corretation between the verhat
total and GRAPH was greatest in FE. The figural total and GRAPH had the
largest correlaLion in MIN.
Regression Analyses on Outcome
Cronbach and Snow (.1977) recommended generalized regression analysts
for investigating ATI. The model for the present study took the form:
Y= 0 t'SS N)131 BG C "T T "ST ST
BDT
DT +BGT
CT
where:
Y = uevendent variable
B0 = constant term
Bs, BD, Bc = regression coefficients for SUM, DIFF, GRAPH
S, D, G = score on SUM, DIFF, GRAPH
T= regression coefficient for treatment
T = orthogonal treatment contrast
BST'
BDT'
$GT
= regression coefficients for first-order ATI
Two orthogonal contrasts represented the three treatments. The
first contrast (T1) compared MIN with VE and FE (coded 2, -1, -1,
21
1 ',1/1.111 1 \'4.,1 1 I II.: 4241 "41,1 , 11114 ,1421 I to
I I cz, I I v 4: I v I h ^I I n 11,Ic11 Iil 1 4111110 WO 1-, 111141111 42.1 le V 111111 1 I
III V 1111; 11 011111142 :111 1 11111 J I 1 11 :4111 I I 11,1c. ,
4111;11 :I I d 421142 I yZIC:1 tsle 111111111 1 1111 111 1 111 11441 I :11'111 can ilki
ImmodIalc and dolayod 12I 1211 1111 III (t.:till( 43 WIWI 11:141(1, W1 1 11
VIII 1 lib 14201 turf vd 11110 ( 1142 oqnat lion III a a11c11 l r 1041 1111101 . Ati( 1111110 ma In
r Za W4-11.0 Oili t! 10(1 1) 1)1 11 d41( (HOU! r 4,21 1 d, Wi I Il ;1111 rani oral! r I I d1 I I)
I Or* I IIto HuttelaI --allI t It y hypot DiFF was ent tared next to test the
a t 101'eut hit -011)1 1,1 yitnthwa CHAN! wita I lion out rtat 141 Atititulti Itti
specific contribution, independent ot the proportion or its effects that
were associated with general and differential ahilify. Treatment main
effects were then entered using the two orthogonal. A"ontranta. Following
all main effects, first-order ATI were entered. Thin order wan used for
all dependent varlables.
Total Score ATAlyses
The results of the regression analysen for the !mmediate and delayed
total posttests are presented Lu Table 6. The percentage of variance
accounted for reflects the change in the squared multiple correlation
coefficient as each predictor entered the equation. The F-ratio tested
whether this change in R2was significant using the formula:
AR2/Ki
1-R2
t/N-K
t-1
where:
AR2= increment in R2
K = number of predictors in change
R2
t= R
2for full model
N = total sample size
Kt= number of predictors in total R2
22
I (.
',Willi,' I ', 61 I. 111/ 114. k I' 4'f1,1 hilt I I I..I.1I 1%1 ,f 1
V i I I'
Fill I 1141,14. I
A111 1111411' 1'141 1114 I tri
II I
I I
1111114'.1 I :11 rti::1
VZi I I
114 11411111 4'41 1.'141
')
/
.3
1:1i III
?1..11
lII.'I
V:I I 141111.4!
Acwint
611.6
I .11
1.111A
II 1.1/11A
!111t1 '11. 1 ??1{. /11A (11 . 1 }OA
III VI' .4 I.ItI .1
(:1(A1'11 11.11 4. 'IBA 1. 'I 14.1)A
411 Iwo! Ha III 1...1 I vt't ;I /1. () 4). t`ok 2. /
1'1 4.0 16. 2',)A I .11 ().(18,1/4
2 .1) 2.44 .1) 1,44
IC:11 01414'4' Al' I It 1. 1 2.24A .1
SUM x '1'1 2.9 I . /BA .0 1SUM X '12 I. .0 <I .0 <1
DIFF X 'I'I. .3 I . 22 .2 1
DIFF X 12 .0 .1.
GRAPH X Ti .0 .0GRAPH X '['2 .1 .4 1.53
Residual 120 29.5 31.4
.05
.WI I1.!.1t. .....,i i 1 .iii.I . t Ii .1
I ht.; I .II 1.t. I I ti . ,I, I I I; 1.14" .. H. I l 13,,1% 11 I II
A...:1...,1 p.,,,11,_,t., 111.w ILI 1 I i It II.IL:Al L I 1,,,,, ,,,11 IL". i i, i, IL.' . .1 , I 1, ,
,
III (110 CI III 111111ll,t1 I AI 1.11:1 I i C:31 .111t1 *,11 hi: I I
II 110 tit I I 0:11
lilt III t I I LI, I 71 1111411 L!.I l "I I. sif I It 11 I tit, c I ti I ht.,
ImmvdInto iinnt(cI And v41 IAnAc, In 1110 doInvod
Thetiv Iho mcAn dIIIotoinctrs IdonIIIIod In IAhIn
VE And FE Ahowed higher Immodtato outcomel (14n did NIN: Itlioton IA
hotwoon VF. And r. were not lignill Ant . In conllaat, on tho dlayed
11w1 t1 1 4I, MIN lIwwvd .1 highor Average than did VI'. and rr.
Inter:14114mA wery toatod in the general(:'Ott modol, then, nnatand
ardtzed rogronalon cootticionta within each tloatment wort' plottod
grAphivaiiy tot. itorproto,tion, !;IIM X 11.0AiIIWIlt lutel'Ac1 1011 war{
statistically significant on the immediate posttest, but not on the
delayed posttest. This interaction accounted for 2.97. of the hmnediate
posttest variance; it is shown graphically in Figure 1. The relative
advantage of high ability students was most pronounced in MIN. In other
words, VE and FE appeared particularly helpful for low ability students,
reducing the difference between them and high ability students. There
were smaller treatment differences at the mean and reduced ATIs on the
delayed posttest.
Part Score Analyses
Verbal items. The results of the generalized regression analyses
of verbal outcome measures appear in Table 7. Again, SUM accounted for
24
I), 1
i I
lJ
1.
!'l
1(1
0 I l _1 1 I _1
-4.0 1.0 .2.0 1.0 0 1.0 2. 1.0 4.0
YIN
!t1
0 1 I 1
4.0 1.0 '1.0 1.0 o 1.0 1.0 1A1 4.0
ligury I. Relation of Total Posttest Scoren to SUM,with Unatandardized Regression CoefficientsShown in rarentheses.
IIM
Table 7
Summary of Stepwise Regression of Verbal Items
Immediate Posttest Delayed Posttest
Variable d.f. % VarianceAccounted For
F - ratio% VarianceAccounted For
F - ratio
Full Model 11 57.5 14.74* 40.4 7.40*
Aptitude Main Effects 3 41.7 39.22* 34.6 23.23*
SUM 1 37.1 104.68* 34.0 68.48*
DIFF 1 .0 <1 .2 <1
GRAPH 1 4.6 12.98* .4 <1
Treatment Main Effects 2 10.0 14.11* 2.9 2.92
Ti 1 2.8 7.90* 2.9 5.84*
T2 1 7.2 20.31* .0 <1
First-Order ATI 6 5.8 2.73* 2.8 <1
SUM X Tl 1 3.4 9.60* .0 <1
SUM X T2 1 1.3 3.67 .1 <1
DIFF X Ti 1 .7 1.98 .0 <1
DIFF X T2 1 .0 <1 .0 <1
GRAPH X Tl 1 .1 <1 .4 <1
GRAPH X T2 1 .3 <1 2.3 4.63*
Residual 120 42.5 -- 59.6 --
*P. < .05
the largest proportion of explained variance in each posttest. GR.=
accounted for a significant proportion of variance in the immediate
posttest, accounting for 4.6% of the variance in it. Students with
higher aptitude scores had higher outcome scores. DIFF was never signi-
ficant.
Treatment main effects correspond to the mean differences shown
previously. On the immediate posttest, treatment main effects accounted
for 10.0% of the variance; VE was superior to MIN and FE. Treatment main
effects accounted for 2.9% of the variance in the delayed posttest, with
MIN superior to \ and FE.
First-order ATI accounted for 5.8% of the variance in the immediate
posttest; ATI with SUM accounted for most of this variance. Figure 2
shows that VE was particularly helpful for low ability students.
The only significant ATI on the delayed posttest was between GRAPH
and treatment. As shown in Figure 3, only in FE did students with high GRAPH
scores outperform students with low GRAPH scores.
Figural Items. Table 8 presents the results of the generalized
regression analyses for figural items. As before, SUM and GRAPH accounted
for a significant proportion of the variance on both the immediate and
delayed posttests. DIFF was not significant, accounting for only .5% of
the immediate posttest variance and .0% of the delayed posttest variance.
Treatment main effects accounted for 6.6% of the variance on the
immediate posttest and 6.0% of the variance on the delayed posttest.
Thus, on the immediate posttest, VE and FE were superior to MIN, on
average. FE had a slight advantage over VE. In contrast, on the delayed
posttest, both MIN and VE were superior to FE.
27
15
10
0
-4.0 -3.0 -2.0 -1.0 0 1.0 2.0 3.0 4.0
SUM
15
0-4.0 -3.0 -2.0
Figure 2. Relation of Verbal Items to SUM, withUnstandardized Regression CoefficientsShown in Parentheses.
jUE'
-1.0 0
SUM
1.0 2.0 3.0 4.0
VE (.04)
MIN (.12)
FE (.13)
0 5 10 15 20
GRAPH
25 30
15
f-cn
10
FW
0
5
0
0 5 10 15 20 25 30
Figure 3. Relation of Verbal Items to GRAPH withUnstandardized Regression CoefficientsShown in Parentheses.
A I1
GRAPH
Table 8
Summary of Stepwise Regression of Figural Items
Immediate Posttest elayed Posttest
Variable d.f.% VarianceAccounted For
F - ratioVariance
Accounted ForF - ratio
Full Model 11 54.3 12.94* 62.2 17.96*
Aptitude Main Effects 3 43.3 37.86* 54.2 57.45*
SUM 1 35.7 93.65* 49.9 158.66*
DIFF .5 1.31 .0 <1
GRAPH 1 7.1 18.62* 4.3 13.67*
Treatment Main Effects 2 6.6 8.66* 6.0 9.54*
T1 1 5.1 13.38* 2.9 9.22*
T2 1 1.5 3.93* 3.1 9.86*
First-Order ATI 6 4.3 1.88 2.0 1.06
SUM X Ti 1 .8 2.10 .0 <1
SUM X T2 1 1.9 4.98* .0 <1
DIFF X Ti 1 .4 1.05 .6 1.91
DIFF X T2 1 .o <1 .0 <1
GRAPH X Ti 1 .1 <1 1.4 4.45*
GRAPH X T2 1 1.1 2.89 .0 <1
Residual 120 45.7 37.8
*2. < .05
The SUM X treatment interaction was, again, significant on the
immediate and not on the delayed posttest. In Figure 4, the irmediate
regression slope on SUM was shallower in FE than in MIN and VE. FE was
particularly helpful to low ability students in reducing differences
between them and high ability students. The mean disadvantage on the
delayed posttest for students in FE can also be seen.
Interactions with GRAPH accounted for 1.2% of the variance in the
immediate posttest, and 1.4% of the variance in the delayed posttest.
While not significant in the immediate posttest, the interaction suggested
that FE reduced the advantage of students with high GRAPH scores. On
the delayed posttest, both VE and FE reduced the advantage of students
with high GRAPH scores. These relations are shown in Figure 5.
Probleas. As with all other dependent variables, SUM accounted for
the largest proportion of explained variance in the problems on both the
immediate and delayed posttests (see Table 9). Although GRAPH was not
a statistically significant predictor of the immediate posttest, it
accounted for 1.3% of its variance, and did account for a significant
proportion of variance (4.9%) on the delayed posttest. Again, the effects
of DIFF were small. No treatment main effects were significant, and the
only significant ATI was, again, with SUM on the immediate posttest. As
shown in Figure 6, differences between high and low ability students were
greatest in MIN.
Summary of Regression Analyses
SUM accounted for the vast majority of variance in all dependent
measures, and GRAPH accounted for significant proportions of variance in
most. Again, because GRAPH and SUM were correlated, and because SUM was
entered into the regression analyses before GRAPH, effects associated
31
15
0-4.0 -3.0 -2.0 -1.0 0 1.0
SUM
2.0 3.0 4.0
15
Hy
HW
OI... 10
5
FE (.94)
0
-4.0 -3.0 -2.0 -1.0 0 1.0 2.0 3.0 4.0
SUM
Figure 4. Relation of Figural Items to SUM, withUnstandardized Regression CoefficientsShcwn in Parentheses.
A "16
15 15
10
5
FE (.04) MIN (.27)
MIN (.16)
10
5
0 0
10 15 20 25 30
GRAPH
VE (.09)
FE (.10)
0 5 10 15 20 25 30
GRAPH
Figure 5. Relation of Figural Items to GRAPH, withUnstandardized Regression CoefficientsShown in Parentheses.
43
Table 9
Summary of Stepwise Regression of Problems
Immediate Posttest Delayed Posttest
Variable d.f.% VarianceAccounted For
F - ratio% VarianceAccounted For
F - ratio
Full Model 11 45.8 9.21* 58.2 15.21*
Aptitude Main Effects 3 42.0 30.98* 56.3 53.64*
SUM 39.9 88.28* 51.2 147.12*DIFF 1 .8 1.77 .2 <1GRAPH 1.3 2.88 4.9 14.08*
Treatment Main Effects 2 1.0 1.11 .2 <1
Tl 1 .8 1.77 .0 <1T2 i .2 <1 .2 <1
-First-Order ATI 6 2.9 1.07 1.9 <1
SUM X Ti 1 2.1 4.65* .2 <1SUM X T2 1 .1 <1 .3 <1DIFF X Ti 1 .0 <1 .0 <1DIVE X T2 1 .1 <1 .9 2.59GRAPH X T1 1 .5 1.11 14 1.15GRAPH X T2 1 .1 <1 .1 <1
Residual 120 54.2 41.8
*2.< .05
0
-4.0 -3.0 -2.0 -1.0 0 1.0 2.0 3.0 4.0
SUM
0
-4.0 -3.0 -2.0 -1.0 0 1.0 2.0 3.0
SUM
Figure 6. Relation of Problems to SUM, withUnstandardized. Regression CoefficientsShown in Parentheses.
with GRAPH reflect its specific variance, not variance shared with sun.
The differential-ability hypothesis, tested by DIFF was never significant.
Treatment main effects were significant for the total, verbal, and
figural scores on both posttests. The immediate posttest showed a gen-
eral advantage for VE and FE over MIN. VE was most advantageous on
verbal items, FE on figural items. But this trend was reversed on the
delayed posttest where students in MIN outperformed those in VE and FE.
In fact, the lowest mean figural part score occurred in FE.
All significant ATI on the immediate posttest involved general
ability (SUM); VE and FE reduced the advantage of high ability students.
This advantage was reduced most in VE on verbal items and in FE on figural
items.
Similar relations were not found on the delayed posttest. High
ability students continued to outperform low ability students on the
delayed posttest; the relative advantages for students in VE and FE were
not retained over time. Significant ATI were obtained only with GRAPH.
Thus, in this study, instruction that was most effective for im-
mediate learning was not most effective in the long run. ATI effects
suggested that this shift may have come primarily from low ability students
who did not retain the additional information that enhanced immediate
learning.
36
CHAPTER 4
SUMMARY AND CONCLUSIONS
The present study examined the effects of verbal and figural sup-
71enents on learning, and their relation to general abi-ity, Gc and
Gfv, and a specific graph-processing test. This chapter summarizes
prior research and the procedures and results of the present study.
Implications for future research and educational practice are discussed.
The Research Problem
Although numerous studies have investigated interactions between
aptitude and verbal and figural instructional supplements, few consistent
relations have been established. In some studies, the regression of
outcome on aptitude was steeper when aptitude and instructional condi-
tion were matched; in others, the regression was shallower. Some studies
obtained similar regression slopes. These relations varied, in part,
as a function of the delay between instruction and the posttest. Incon-
sistencies also resulted from not specifying aptitude and treatment in
sufficient detail or considering all the ways they might be matched.
The Present Research
The present study compared the effects of minimal instruction,
instruction elaborated with verbal supplements, and instruction elaborated
with figural supplements. Aptitude was represented by a Gc composite,
a Gfv composite, and a graph-processing test. The Gc and Gfv composites
were summed to indicate general ability and their difference was used
to investigate their differential impact. Immediate and delayed outcome
measures included verbal items, figural items,ard froblems that could be
solved either verbally or figurally. Learning was described as a function
of aptitude and instructional material.
37
t- 0
The Effects of General Ability
On the immediate posttest, students in VE and FE did better than
students in MIN, suggesting that the elaboration provided in these con-
ditions helped students learn. Significant ATI suggested that the elabo-
ration was particularly useful to low ability students. High ability
students did as well or better in MIN. Thus, the regression of achieve-
ment on general ability was steepest in MIN and reduced in VE and FE.
Partitioning of the total test score by item type indicated that
VE was particularly helpful on verbal items and FE was particularly helpful
on figural items. Again, significant ATI indicated that these treatments
were particularly helpful to low ability students. Hence, the regression
of verbal items on general ability was least steep in VE; the regression
of figural items on general ability was least steep in FE.
Examination of, learning outcomes on retention, however, led to
strikingly different conclusions. While students in MIN were worst on
average achievement on the immediate posttest, they peiformed best on
the delayed test.
VE and FE provided more information to learners through additional
explanations and examples. MIN required students to provide this infor-
,mation for themselves, thereby demanding more active work from learners.
More able students, capable of doing it, did equally well in MIN as in
VE and FE. Less able students that could not generate that information
for themselves benefitted from the assistance. The gains were short-lived,
however. Active mental work, necessary in MIN, appeared to aid retention.
Hence, there cas a greater decline in performance in VE and FE than in
MIN. This contention was further supported by the observation that losses
38
were greatest when the assistance was most direct. That is, lossos on
verbal items were greatest in VE; losses on figural items were greatest
in FE.
The Differential Effect of Gc and Gfv
The differential impact of Gc and Gfv as measured in this study,
did not enter differently into outcome on aptitude relations. This may
be due, in part, to a failure to adequately distinguish Gc from Gfv.
That is, measures of Gc and Gfv shared a considerable proportion of
variance. Reducing that overlap might increase the chance of detecting
differences in their impact. Thus, future research in this area must
strive to do this.
The Effect of GRAPH
The instructional materials in this study used many graphic displays.
Therefore, GRAPH was included as a specific-ability measure. Sianificant'
main effects were associated with GRAPH at both testings, even after account-
ing for the effects of general ability. As with general ability, students
with higher aptitude scores did better. Thus, learning outcomes were not
fully described by the effects of general ability.
Conclusions and Implications
This study provided evidence that neither aptitude nor instructional
treatment alone can fully describe learning outcomes. Interactions between
them exist and can be demonstrated. Further, instructional supplements,
whether verbal or figural, can be effective in filling-in for students
weaknesses and reducing differences between high and low ability students.
Such supplements, however, must be used with caution. Reducing the
difficulty of instructional materials may indeed enhance immediate learning,
39
but these advantages may be short-lived. In the present study, increasing
the difficulty of the work required for initial learning appeared to in-
crease retention. The benefits in immediate learning must be weighed with
the need to ensure that information is retained.
The implications for educational practice are clear. The present
study indicated that instruction that enhances immediate learning is not
necessarily best for retention. Ultimately, educators must be concerned
with how much information is retained and not limit their concerns to
immediate outcomes. Thus, not only must achievement be assessed at more
than one point in time, curricula must be developed to promote long-term
learning. The current emphasis on testing after only short delays should
be reconsidered.
Additional research is necessary to confirm the findings of this
study and support these contentions. ATI research is one avenue for ex-
ploring this area, but it should be supplemented with more basic research
in information processing. Improved methods of distinguishing Gc from Gfv
are required to explore their differential impact on learning. These
methods may emerge as we gain a better process understanding of these
abilities through further research.
Finally, researchers should examine both immediate and delayed out-
comes, and attempt to identify instructional conditions likely to promote
long-term retention. At a minimum, delayed outcome measures may be added
to instructional research conducted in different contexts. Additional,
more directed research may probe more deeply'the relations among instruc-
tional materials, immediate learning, and retention. The present study
suggests that we cannot limit research to immediate outcomes if we are
truly interested in the long-term impact of instruction.
In conclusion, this study examined the relations among aptitude,
instruction,,and learning. While it provided data to help answer some
questions in this area, it raised many others that only future research
may resolve.
41
References
Allison, N. B. Learning parameters and human abilities. Unpublished
report, Educational. Testing Service, 1960.
Bracht, G. H. Experimental factors related to aptitude-treatment inter-
actions. Review of Educational Research, 1970, 40, 627-645.
Butcher, H. J. Human intelligence: Its nature and assessment. New York:
Harper and Row, 1968.
Carry, L. R. Interaction of visualization and general reasoning abilities
with instructional treatment in algebra. Unpublished doctoral disser-
tation, Stanford University, 1967.
Cattell, R. B. Theory of fluid and crystallized intelligence: A critical
experiment. Journal of Educational Psychology, 1963, 54, 1-22.
Cattell, R. B. Abilities: Their structure, growth, and action. Boston:
Houghton Mifflin, 1971.
Cronbach, L. J. The two disciplines of scientific psychology. American
Psychologist, 1957, 12, 671-684.
Cronbach, L. J. and Snow, R. E. Aptitudes and instructional methods: A
handbook for research on interactions. New York: Irvington, 1977.
Eastman, P. M. The interaction of spatial visualization and general reas-
oning abilities with instructional treatment in quadratic inequalities:
A follow-up study. Unpublished doctoral dissertation, University of
Texas, 1972.
French, J. W., Ekstrom, R. B., and Price, L. A. Kit of reference tests
for cognitive factors. Princeton, New Jersey: Educational Testing
Service, 1963.
Graham, F. K. and Kendall, B. S. Memory-For-Designs Test. Perceptual
Motor Skills, 1960, 11, 147-188.
42
Guilford, J. P. The nature of human intelligience. New York: McGraw-
Hill, 1967.
Gustafsson, J.-E. Verbal versus figural in aptitude-treatment interactions.
Review of the literature and an empirical study. Reports from the
Institute of Education, University of GOteborg, No. 36, 1974
Gustafsson, J.-E. Verbal and figural aptitudes in relation to instructional
methods: Studies in aptitude-treatment interaction. GOteborg Studies
in Education Sciences 17, GOteborg, Sweden, 1976.
James, M. A. and Knief, L. H. Interaction of general, fluid, and crystall-
ized ability and instruction in sixth-grade mathematics. Journal of
Educational Psychology, 1978, Vol. 70, No. 3, 319-323.
Lipsey, R. G. and Steiner, P. 0. Economics. New York: Harper and Row,
1969.
Lohman, D. F. Spatial ability: A review and reanalysis of the correla-
tional literature. Technical Report No.8, Aptitude Research Project,
School of Education, Stanford University, 1979.(a)
Lohman, D. F. Spatial ability: Individual differences in speed and level.
Technical Report No. 9, Aptitude Research Project. School of Educa-
tion, Stanford University, 1979. (b)
Markle, N. H. Differential response to instruction designed to call upon
spatial and verbal aptitudes. Unpublished doctoral dissertation,
Stanford University, 1969.
Peterson, J. C. and Hancock, R. R. Developing mathematical materials
for aptitude-treatment interaction. Paper presented to the American
Research Association, 1974.
Raven, J. C. Advanced Progressive Matrices. Great Britain: Silver End
Press, 1962.
3
t I I 1111 t Honristio modols for the generation of aptitude-treatment
Interaction hypotheses. Review of Educational Research, 1972, 42,
327-343.
Samuelson, P. Economics. New York: McGraw-Hill, 1976.
Snow, R. E. Research on aptitudes: A progress report. Technical Report
No. 1, Aptitude Research Project. School of Education, Stanford
University, 1976.
Snow, R. E. Individual differences, instructional theory, and instructional
design. Technical Report No. 4, Aptitude Research Project, School
of Education, Stanford University, 1977.
Snow, R. E. Aptitude processes. In Snow, R. E., Federico, P.-A., &
Montague, W. E. (Eds.) Aptitude, learning, and instruction: Vol. I,
cognitive process analyses of aptitude. Hillsdale, New Jersey:
Erlbaum, in press.
Snow, R. E., Marshalek, B., Lohman, D. F., Yalow, E. S., Webb, N.
Correlational analyses of reference constructs. Technical Report
No. 5, Aptitude Research Project. School of Education, Stanford
University, 1977.
Spencer, M. Contemporary economics. New York: Worth, 1977.
Sutton, H. Contemporary economics. New York: Praeger, 1976.
Terman, L. H. Concept Mastery Test. New York: Psychological Corporation,
1956.
Vernon, P. E. The structure of human abilities. London: Methuen, 1950.
Webb, L. F. Interaction effects between selected cognitive abilities
and instructional treatment in algebra. Unpublished doctoral disser-
tation, University of Texas, 1971.
Wechsler, D. Wechsler adult intelligence scale. New York: Psychological
Corporation, 1955.
44
APPENDIX A
MEANS, STANDARD DEVIATIONS, AND CORRELATIONS BETWEEN APTITUDES-
Meas, St audit t'd 'ley t. , awl Cot' rt... I at lam.
of AptItude Measures In MIN (N " 44)
Variable Mean S.D. (I.) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
(I) Vocabulary Multiple-Choice 15.20 4.83 71 64 12 02 22 31 42 89 22 65 67
(2) Vocabulary F111-In 27.18 5.15 65 23 25 33 39 57 39 74 49
(1) 'Ferman Concept Mastery 30.64 5.29 49 41 49 69 68 87 00 20
(4) Copylng 32.36 9.09 58 60 44 32 85 68 -52
(5) Memory for Designs 17.16 L.90 37 15 24 26 63 51 -36
(6) Paper Folding 13.80 3.66 53 34 39 82 70 -42
(7) Advanced Progressive Matrices 23.59 5.15 66 53 75 74 -21
(8) GRAPH 22.11 5.95 63 55 68 09
(9) Cu .08 .99 49 86 51
(10) Gtv .06 .96 86 -50
(11) SUM .14 1.f9 01
(12) D1FF .02 .99
Note. Decimals omitted from correlations
MOi110, ;;iandard neviailoun. and Corielallona
61 Apt I t udc tioanuteu In VI.: (N ti/;
Variable Mean ;;. I). (I) ( 2) ( (4) (6) (/) (11) (9) (10) (II) (12)
(I) Vocabulary MullIplo-Choice 15.25 4.73 61 66 26 1.6 36 33 14 85 :16 ,68 52
(2) Vocabulary F111-in 26. 14 5.89 16 48 57 61 49 91 66 8/1 25
(1) Terman Concept Mastery 30.16 5./4 25 29 50 50 43 91 51 80 43
(4) Copying 31./5 9.03 33 62 57 54 37 81 67 -49
(5) Memory for De!;igns 16.98 2.02 34 40 51 30 66 54 -40
(6) Paper Folding 1.3.43 1.85 50 64 54 81 76 -31
(1) Advanced Progressive Matrices 24.43 5.99 62 55 81 77 -30
(8) GRAPH 23.48 5.51 41 75 65 -39
(9) Cc .01 1.03 58 89 44
(10) Cfv .03. 1.05 89 -48
(11) SUM .04 1.85 -02
(12) D1FF -.02 .96
Note. Decimals omitted from correlations
Means, :;tnudald nevintlonn, snd Cottelntios.,
tot 1\1.1 I I kith, 14. (II
V,1r I,IIile Mean !;.D. (I) ("7.) ( /) (4) (6) (1) (8) () ( 10) (11) (12)
O) Vocahnlary MnItIple.ChnIce 15.48 4.1? 1(1 /9 49 15 1/ 52 46 9'2 80 41
(7) VocahnIary FIII-In 2/.16 t9 17 I)1 50 I) ) Ysl /8
(0 Teman Concept Mastery 29.91 4.14 /19 21 1/ 9t) /9 40
(4) CopyIng 11.64 10.82 /1/4 40 53 11 48 /9 73 -'16
(5) Memory For Designs 11.14 2.01 26 59 38 20 73 54 -58
(h) Paper Folding 11.32 3.15 61 44 46 11 67 -29
(/) Advanced Progressive Matrices 25.52 5.25 56 56 87 82 -36
(8) CRAPH 23.05 4.99 56 54 63 00
(0) Gc .05 .97 55 87 44
(10) Gfv -.01 1.02 88 -51
(11) SUM .04 1.75 -05
(12) DiFF .07 .95
Nolo. Decimals omitted from correlations I I'
I X It
POSTTI LAT 1
49
V:11 I
1111114'11 1,11 a 11 1. ,1
I 1 1. 1 1
Iii) ! 1..11 I 1.11
11 I It,111 I 1,,i I
( 7i) l'i ,I) I
I
NI)t. 4. III.,- OW It t 141 I I (.in r 1,1 1 4 IA( I ow; .
1(cl 1.111 lit( I .11)p.11 III 1114 111.1111 ti
I
I t
i
t, I
it'.
',/
Posttest. Correlations in MIN (N =
Variable (1) (2) (3) (4) (5)
Immediate Posttest
(1) Total Score 86 89 84 92
(2) Verbal. Total 66 56 87
(3) Figural Total 64. 80
(4) Problems
belayed Posttest
(5) Total Score
(6) Verbal Total
(7) Figural. Total.
(8) Problems
70
Note. Decimals omitted from correlations.
44)
(6) (7) (8)
75 81 82
75 73 78
61 81 62
57 56 71
82 88 89
56 65
67
Variable
Posttest Correlations in VE (N = 44)
Immediate Posttest
(1) Total Score
(2) Verbal 'Total
(3) Figural Total
(4) Problems
Delayed Posttest
(5) Total Score
(6) Verbal Total
(7) Figural Total
(8) Problems
0 (2) (3) (4) (5) (6) (7) (8)
75
-
86
48
-
78
40
48
-
87
65
78
63
69
65
59
42
87
-
78
50
78
54
93
76
83
61
68
71
84
58
65
Note. Decimals omitted from correlations.
Variabl
Ineuediate
(1) To
(2) Ve
(3) Fi
(4) Pr
Delayed Po
(1) To
(2) Ve
(3) Fi
(4) Pr
Posttest Correlations in FE (N = 44)
(1) (2) (3) (4) (5) (6) (7) (8)
'osttest
:al Score 81 72 8(1 88 71 76 75
:bal. Total 41 44 71 67 60 54
;oral Total 36 67 44 70 54
)blems
ittest
67 54 50 66
Ca 1 Score 80 88 84
r b a 1 Total - 57 50
;oral Total 62
)blems
Note. Decimals omitted from correlations.
APPENDIX C
CORRELATIONS BETWEEN APTITUDE AND OUTCOME
Vorabulai
Vocabulat
Turman Cc
Copying
Memory (
Paper Fo
Advanced
GRAPH
Cc
Gtv
SUM
NIT
Correlations Between Aptitude and Outcome
in Total Sample (N = 132)
Immediate Posttest Delayed Posttest
Total Verbal Figural Prob- Total Verbal Figural Prob-Score Total Total lems Score Total Total . lems
-y Multiple Choice 52 41 39 48 51 39 44 49
-y Fill-in 63 !(.) Yi 54 (6 53 513 59
,ncept i!astery 6R 57 c7 69 49 63 65
41 40 26 35 46 30 44 42
)r Designs 40 40 35 21 42 36 35 38
ding 49 34 36 49 54 36 53 49
Progressive Matrices 64 51 51 51 65 49 59 58
66 56 58 49 65 42 61 63
69 53 56 60 70 53 62 65
63 54 49 51 68 49 62 61
75 61 60 63 78 58 71 72
05 -02 06 08 01 03 -02 03
Note. Decimals omitted from correlations
Vocabulary h
Vocabulary I
Terman Conc.
Copying
Memory for
Paper FoldirIn
Advanced Pr(
GRAPH
Gc
Gfv
SUM
DIFF
Correlations Between Aptitude and Outcome
in MIN (N = 44)
Immediate Posttest Delayed Posttest
TotalScore
VerbalTotal
FiguralTotal
Prob-lems
TotalScore
VerbalTotal
FiguralTotal
Prob-lems
lultiple Choce 53 42 40 56 47 34 38 50
'ill-in 62 52 53 57 62 44 53 64
pt Mastery 79 68 67 70 73 51 64 74
50 57 32 40 45 24 47 44
)esigns 49 60 41 23 50 50 40 40
ig 48 39 40 47 48 28 48 48
)gressive Matrices 68 52 63 62 60 35 59 59
69 63 65 50 71 38 76 65
73 61 60 69 69 49 58 71
70 67 57 57 66 I:4 63 62
83. 74 68 73 78 54 70 77
04 05 04 13 03 05 04 09
Note. Decimals omitted from correlations.
Correlations Between Aptitude and Outcome
in VE (N = 44)
Immediate Posttest Delayed Posttest
TotalScore
VerbalTotal
FiguralTotal
Prob-lems
TotalScore
VerbalTotal
FiguralTotal
Prob-- lems
Vocabulary Multiple Choice 45 32 39 35 50 48 43 43
Vocabulary Fill-In 74 50 64 62 76 71 67 65
Terman Concept Mastery 63 38 59 51 71 54 68 65
COpying 44 30 43 31 53 52 50 39
Memory for Designs 38 21 46 19 38 28 41 29
Paper Folding 54 33 39 58 62 47 63 52
Advanced Progressive Matrices 64 51 61 40 71 67 67 55
GRAPH 64 33 64 47 62 42 61 58
Gc 68 45 61 56 75 65 67 65
Gfv 66 44 61 49 73 63 72 //
57
SUM 75 50 69 59 83 72 78 69
DIFF 02 00 -02 06 00 01 -07 07
Note. Decimals omitted from correlations.
Vocal
Vocal
Term
Copy
Memo'
Pape
Adva
(;RAP
Cc
Ctv
SLIM
NIT
Correlations Between Aptitude and Outcome
in FE (N = 44)
Immediate Posttest Delayed Posttest
TotalScore
VerbalTotal
FiguralTotal
Prob-lens
TotalScore
VerbalTotal
FiguralTotal
Prob-lems
milary Multiple Choice 66 57 /6 50 60 40 56 54
)ulary Fill-in 71 66 51 48 65 48 6/: 51
in Concept Mastery 71 59 55 51 63 44 58 57
Lug 38 39 08 37 41 20 40 42
ry for Designs 40 47 20 23 41 30 29 44
r Folding 55 43 40 45 54 35 49 50
Iced Progressive Matrices 64 58 38 52 67 51 . 56 61
I65 65 19 46 70 58 54 66
76 68 55 55 70 49 66 59
62 60 33 50 64 43 56 63
79 72 50 60 76 52 69 70
12 05 22 03 03 04 08 -07
Note. Decimals omitted from correlations.
....au, as .Nlubr ',Mee-Muer ju 140,,
Navy Navy
rage
Navy
atiOve yecem se, j,r,
Navy
rage e
1 Meryl S. Baker Dr. Patrick P. Harrison 1 Dr. George 'Moeller 1 Orrice of Naval ResearchNPROC Psychology Course Director Mead. Hum an Factors Dept. Code 441
Naval Submarine Medical Research Lab 600 N. Quincy StreetCode P309 LEADERSHIP & LAW DEPT. (71a),Cdt_06 340 Arlington, VA 22217San Diego, CA 92152 DIV. Car PROFE.SSIONAL DEVELOPMMENT Groton,
U.S. NAVAL ACADEMY1 Dr. Robert EVeau ANNAPOLIS, MD 21402 1 Dr Willia lbm 'ntague 5 Personnel II Training Research Programs
Code N-711 Navy Personnel R&D CenterSan Diego, r 0 '152 Office o!NAVTRAEWIPCEN I Dr. Jim Hollan Research
Orlando, FL 32813 Code 304 Arlington, VA 22217Navy Personnel R & D Center 1 Loalbvaralrney
alt` ,e arch Center 1 Psychologist1 Chief of Naval Education and Training San Diego, CA 92152Liason Office P. 0. Bo: 85'22 ONR Branch Office
Air force Murton Resource Laboratory 1 CDR Charles W. Hutchins San Diego, CA 921 38 1030 East Green StreetFlying Trainirg Division Naval Air Systems Command I1 Pasadena, CA 91101WILLIAMS AFB. AZ 85224 AIR-340F 1 Naval Medical R4D Command
Navy Department 1 Office of the Chief of' Naval Operations1 Or. Larry Oran, LT, MSC, USN Washington, DC 70361 NationaltionCall Naval Medical Center Research Development & Studies Branch
Bethesda. MD 2001 4Psychology DepartmentWashin(OgtoP-n11,5DC) 20350Naval Submarine Medical Research Lab CDR Robert S. Kennedy
Naval Submarine Base Head, Mangan Performance Sc iences Ted M. I. Yellen05340 Naval AerospaceGroton, CT rospace Medical Research Lab Technical Information Office, Code 201 1 Dr. Donald F. Parker
Pox 29407 NAVY PERSONNEL 114D CENTER Graduate School of P,ainess AdminiatratiSAN DIEGO, CA 92 152 University of 111, , ..11 Dr. Richard Cater New Orleana, LA 70189
Departmeet or Administrative Sclencea Ann Arbor, MINaval Postgraduate School 1 Dr. Norman J. Kerr 1 Library, Code P20 11
Navy Personnel PAD Center 1 CL.TdFerLan5k1 C. P- USN (Ph .D)Naval Air Station Memphis (75/
Monterey, CA 93940 Chief of Naval Technical TrainingSan Diego, C4 92152
Naval Aerospace ", ,1 Research LaboratI DR. PAT FEDERICO Millington, TN 380546 Commanding Officer Pensacola, FL 32508NAVY PERSONNEL RIO CENTER ,
Naval Research LaboratorySAN DIEGO. CA 92152 1 Dr, William L. MalayPrincipal Civilian Advisor for fwd e. 2627
Washington, DC 20 390 Operations Research DepartmentI Dr. Gary POOCIC
1 Mr. Paul Foley Education and TrainingNavy Personnel /15O :enter Naval Training Command. Code OJA Cede 55PK,
1 Psychologist Naval Postgraduate SchoolSan Diego, CA 92152 Pensacola. FL 32508ONR Branch Office Monterey. CA 93940
Dr. Jahn Ford 1 Dr. Kneale Marshall Bldg 114. Section D666 Summer Street 1 Roger V. Remington, Ph .DNavy Personnel 1150 :enter Scientific Advisor to DCHOOIPT 1
San Diego. CA 92152 OPOIT Boston, NA 02210 Code 152Washington DC 20370 NAMAL
P1 Psychologist Pensacola, FL 32508Dr. . Henry M. Hal ft'Department of Psychology,C-009 1 CAPT Richard L. Martin, USN OCR Branch OfficeUniversity of California at San Diego Prospective Commanding Officer 536 S. Clark Street 1 Dr. Bernard Rimland (0 3B)La Jolla. CA 02093 USS Carl Vinson (CVN-701 Chicago, IL 6060 5 Navy Personnel R&D Center
Newport Neva 9tipbuilding and Crydock Co San Diego, CA 92152LT Steven D. Harris, MSC, 113.1 Newport News, VA 23607 1 Office of Naval ResearchCod. 6021
Br 8C DO X 1 .d0 30.17
1 Dr. Worth ScanlandQuincy SStreet Chief of Naval Education and Training!Naval Air Development Center 1 Dr. Jame, Mcide
Arlington. VA 22217 Code N-5Werminater. Pennsylvania 18974 Navy Personnel RSD CenterSan Diego. CA 92152 NAO. Pensacola, FL 32508
BEST COI' AVAILABLE C,9L.) tr
3.1
Wavy
I Dr. Sea Schiflett, 30 721Systems Engineer log Test DirectorateU.S. 114161 Air Test. CenterPa turent Ri v er ;/0670
t Dr. Robert G. 7111thOrrice of 011e f of Naval OperationsOF-917HWashing ton, DC 20150
Dr Al fret F. Os,leTrain: rig Alit l ysis I Evaluation Croup
(731)Dept. of toe re y
Or 1 ac 1n, FL R 1 3
1 W. Gar y ThexasonNaval rk, ear Systems:.ode 71327.Ro rtr CA 9235?
sager Wel aslnyer-Fla ylo^re-pir...nent of Adnloist ICI., ''"Navel idit grad oat,- "r.!. ofmonterey, , CA 73740
1 Dr. Ronald 4^1t moancc,f,tepirtwent of LInsiniRtr ativr SciensesD. S. Navel Rostgradote SchknolA,nterry, CA 913AO
1 Dr, Robert WisherCode 337Navy Per 1,rnel RtD Cent,-San Diego. CA 921',2
1 DR. MARTIN F. 14 iSK./F1NAVY RERSVINEL RR 0 CENTERCAN DIEGO CA 73152
r John H. Wol reCafe 111nU. S. Navy ler-sort,: R 1- 1-
Develomen.. cent-rSan Diego. CA -71c2
".
(Ci
1..,
!; :"1
Aruy Air Force
1 Tech/11okt DirectorL. 5, Array Rcaarch to at ituts for ter
(Saha, torsi and S>ofal Se lances5001 El sen1adve r Ay en,,,Al egatalr la VA 22111
HQ USA R g f th troyOK' 501SUSAAREUE Dire-ctor of CLDAPO Maw TorI 09 0 3
DS, RALPH fallSEKU.S. ARMY RESEARCH INSTITUTE5001 EISENHOWER AVENUEALEXANDRIA VA 22311
Dr, Eoa ter FletcherU.S. Army Research Institute5001 ES senho we r Ay enueAl esandr la ,VA 22331
Dr Michael KaplanU.S. ARMY RESEARCH INSTITUTE5001 EISEN/I-NES AVENUEALEXANOR IA, VA 22731
Dr, Milton S. KatzTr ainIrig Technical AreaU.S. Army Research institute5001 El senhower AvenueAl txanclris. VA 22333
1 Dr. Harold F. O'Neil JrAttn: PERI-OKArmy Research Institute5001 Et se.nhower Av enucAlexandria VA 22133
1 Dr Robert SaskorU. S. Army Research institute for the
Dehav lora' onl Social Sc tenors5001 Eisenhower AvenueAlexandria VA 22333
1 Or Josephs WardU.5, Army Research Institut'5001 El senhower AvenueAl eA anle la VA 22331
Air University LlirarpAUL/LSE 76/043Maxwell AFB. AL 36112
. Earl A. Al lui 31110. AFFIRL ( unc)brooks AIR, TX 711235
Dr. Genevieve HaddadProgram 'tanagerLife Sciences DirectorateAF WitLAoIIing AFB. DC 20332
1 Dr. Ronald G. HughesAFIIIL/OTS41111tres AID, AZ 85220
1 fir. fOlcola FeeAFFIRL/HPBrooks RIB, TX 79235
1 Dr. Marty RockwavTechnical DirectorAFYIRLI07141111 isms AFEI. AZ 5522:1
3703 TCITOUTTCH Stop 3?Sheppard AID, TX 76311
1 Jack A. Thorp, Raj., MAELife Sciences DirectorateAFOSRDot line MI. DC 20332
Co astCuard
egi
Chief Psychological Re sorch branchU. S. Cnast Guard (C-P--1/2/TP337)Washingt.r. DC 20593
mr, Thoass A, WarnU. S. 7-oast Guard inst 1 tut-P. 0. Substatio-CRA1.bola City. OK '31E7
Mar Ines
Willlai Cr eenupEducation AdI tsar (EDP'Education Center, NCOECCitontico, VA 22110
Ileedqur ter s, U. S. Mar in' CorpsCoale 1111-201.1shing ton. LC 20360
1Special Assistant for Mar Inc
Corps MattersCode 100MOrrice of Naval Research100 N. Quincy St.kr 1 ingt,n, VA 22217
1 DR. A .L. SL.AFKOSKYSCIENTIFIC ADVISOR (CODE RD-1)HQ, U.S. MARINE CORPSWASHINGTON DC 20360
Let r
2 Defense Technical In format1 o. looterCasieron Statino, Bldg 5Al ex and.rla VA 22311Attn: TC
Mil Story Assistant for Trcln Say gnePer- ,noel Technology
Office L the Under f...cretary of De( ,e
for Re search A Fngineer trigRoca 30125. The PentagonWashington, DC 20301
1 . Susan C111phanLearning afkl C.,elopaentRat 13cal Institute of Edo: at ion1200 19th Street kwWashington. Ca: 20109
1 Dr. Jose pr I. Li psonSERA J-638National So fence FoundationWashington, DC :0550
William J. Pk:LaurinRm. .11, Internal Revenue Sera ice2221 Jeffer son Davis HighwayAr 1 tngton. VA 222912
Dr. Andrew R. MolnarScience Education .
and ResearchNat lanai 7.clence Fotro1.0.1,,nWashington,
Per son,. I RI') Centernrt Pe11,nnel 1.1 tn
197, 5Wa ant ne.t.,r,, L. ;0415
. H. Wallace Sinniko?rogr am DirectorA1.1,1 pow..., 5ea -ch .1 A...1,1n, r 4
7esithsonlan last It Lion901 North Pitt StreetAl e. andr la , VA 22119
Dr, Frank WithrowU. S. ))f face of fdu-at ionandersr yl and An elialhlngt.,,n, DC ;022;
Dr. Jos, .11 L. YoLog , Director0-,nor-y I C.oglitlre4.11.1 :141 Yctencr Foundation
rg tor.. DC 2°561
Non Cdvt
Dr. John R, Ander ionDepartment of P3 yr. hnlogyCarneg le lie11on UniversityPI tt3t,u gh. PA 15211
1 Anderson, Dumas, H., Ph.D.Center for the Study of 53 ad ing174 Children's Research Center51 Gerty Cr lyeChoplagn, IL 61920
1 Dr. John ArnettDepartment. of F'SychologyOnivernity of WarwickCoventry CY4 SALENGLAND
1 DR. MICHAEL ANOODti'E APPL:CATIONS INSTIT !TE
an DENVER TECH. CENTER WEST7915 E. PREYT IC'E AVENUEENGLEWOOD, CO 90110
1 psychological research unitDept. of Defense Army D. rice)Campbell Park Off innCanberra ACT 2600. Aust.- al to
Alan Poddeley`c final Research Council
Appal l Psychology Unit15 Chaucer 11,,m1Cambridge CB? 2EFENGLAND
Dr. Pate Iola DaggettDepartment of PsychologyUn/ of Denverhelve' 31 ty Park
.ver. CO 80209
Aaron PorrpArtnPn,.. of Carardte-r
d.anford UniversityStanford , CA 05105
Non Gov t
1 Dr. Jackson IleattyDepartment of PsychologyUniversity of Cali Corot aLa, Angeles, CA 900211
Isaac BearEducational Testing S.,' Y17,Pr inceton. NJ 03550
Dr. Ina BilodeauDepartment of P3ycha.al:dy1Tolima UniversityNew Orleans, LA 70118
1 Dr. o.chol as A. BondDept. of PsychologySacramento State College600 Jay StreetSacramento, CA 95919
1 Dr. Lyle BourneDepartment of PsychologyUniversity of ColoradoBoulder CO 80309
1 Dr. Robert EY-ennanAmerican Col lege Te st log Pr-oar:zooP. 0. Boa 169Iowa City, IA 52240
1 Dr. Pruce BuchananDepartment of Computer ScienceStanford UniversityStanford CA 95305
1 DR. C. VICTOR BUNC'E RSONWICAT INC.UNIVERSITT PLAZA. SUITE 10
1160 SO. STATE ST.09E11 LIT 85057
BSST 2
71A, 11
J ABLE
Dr. Pat CarpenterDepartment of Psyc'sologyCarnegie-Mellon UniversityPittsburgh, PA 15213
1;.,r1 Gov t
1 Dr. John B. CarrollPayrhoesetr lc LabUniv. of no. ClrolinaDavie Hall 011.Athnpel Hill, NC 2751A
1 Charles Myer:: I 'aryLivingstone Roo,-Livingstone RoadStratfordLondon E15 2LJENGLAND
1 Dr. William ChaseDepartment of PsychologyCarnegie Mellon lln lyre sityPittsburgh, PA 15213
1 Dr. Kenneth E. ClarkCollege of Arts I. SciencesUniversity of RochesterRiver Campus .St ItionRochester NT 14627
Dr. Norman CliffCopt. of PsychologyUniv. of So. CaliforniaUniversity ParkLos Angeles. CA 90007
Dr. Lynn A. CooperI.RDCUniversity of Pitt35,-gh3939 O'Hara StreetPittsburgh, PA 15213
Mere,11th P. CrawfordAm.-icon Psychological Asa:, tat ton1200 17th Street. N.W.Washington, DC 20719
Or. Kenneth N. Crossknocapa Sc iences. Inc.P.O. Drawer C
',a Barbara, CA c31C2
1
Nun
Pe. lion,* id 1hpartwent or 1.1 anon amt IJu
Car e, IL 6."Pi
Dr . fy,:a onto,: nc hi riDepar,nnt Pay-rology
city of Ill Ins!,Champaign, 61021
1 (Jr. lkinert treyfulTe pa r tra-.t of Fllti os.rpnyItn ity or roll( -rnl.Ileraely. CA 9a720
Dr. 411: am ILn14p7-p r Pay:7,01.:1Tulane altiNew 'e; -sn1, LA 7,111
1)1.4!.C7,.--/ ,VVEL A'
101 C,)1, vl1. '13 DR';; 1.1'd A. 71N4 .lA r,4
EA:C t3 ttit.r..A, pit lit I nosa' 1) !s-Li-syN. the a:a , MD :0014
oil L. Fer.ui,n;he Coe -icon College Testtn; ProP . 0 . 159
, 5:240
' 1 non
Dr. 4. riels.oan1.-1v.r.cel RI, 3^ 4,7r1 Cel Cogan.fUlte '.70330 F33'. Wes.. ltigi.,tyWsaningt,n, DC ?O')IA
Dr. Jahn R. Fr ea er (lien!Solt Ber .trek 4 Ac 14,1
'ant, idg e u `,1
*
in dt 111.1r/merit. or Pay.- r
Odiveralty A1',r,sda.,nton Alber
CANADA I' 9
pr R. ari want Or 1 or l a 10De pa r 1m en t l'Ivt,lo4yUn iversi ty of Caltfornlitlos Angeles, CA 91024
R0111111 GLASERLA DC
UNIVERSITY OF PITT5511t1Gd19)9 O'HARA STREET
PITTSBURGH, PA 15:13
. Mary in D. Gluck217 Stone bollCornell 'in ,' t yIthaca, HT 10043
. Doh lel C,;hcrtat I ma- Eng lneer1no.
I s' a,: ?nat. ltute110
r:RA FL
D.O. 204E5 S. 07E54)LAXUNIVERSITY OF PITTSMURCM3939 O'HARA STREETPITTSBURG4. PA 15213
1 Dr. Harold NasratnaDcp.rDrnent of PsychologyUniversity of Dreg:,Eugene OR 97103
Dr . James R. RoffmanDeportment of Fsy:holrgyUa lverstty of Dml sus:,
warlr, DE 19711
Glenda Greenwald Ed
"11Jra an Intelligence wa I etter"P. 0. Box 1163BIrminghms. NI 0,±112
13
1
1,
0 1,
Dr, Jill Larkinof Ps (..partneot of pw:hology
f of III Carnegie Mellon 11t1ver sit Yll 613,, Pitt burgh. PA 15:11
1,1brar y11,01N9/ Western Ill, t.'1047 Berwick fa Iv,.Carmel. CA 91121
Or . Earl !knitDept, of PaynhoIngyUniversity of WashingtonSeati.le, VA 93105
1 Dr. Steven W. ReeleDept. or PsychologyUniversity of OregonEugene, OR T7403
1 Or, Waiter KintschDeportment of Psychologyuniversi ty of W1 or aloBoulder CO ?.1312
1 Dr. Duo Id VieraaDepartment or PAycl,1,,tyUniversity of AdanaTuscan, AZ 95721
Or, Rennth A. lily ins tonFragrant OfficerAlfred P. Sloan Foundation630 Fifth AvenueNew Ycirk, NT 10111
1 Dr. StePnen KosslynMary ard thiversityDepartraent of Psycho tog y33 Kirklanl StreetCambridge, Mk 02133
I Mr. Marl In Fri1117 Ito Cole'Palos Verdes 3. CA 90274
Dr . Alan le 'goldorning RAD Center
On iv,- ty of PittsburghPi Ltabor gh, PA 15260
Dr, Charles LevisFacul te it foe fele laetenachapyr1Aijksun Iveratteit GroningenOule ft" ter inga atraatGroningenNETHERLAND',
Dr. James Ltam mienDepartment or Ps ycho I or; y
tversl ty or Western AustraliaNeill and 3 'W. A, 61109
AUSTRALIA
Dr. Marie HillerCorapdtrr lence Lab. storylean Instruments.Ma II Station 371, P.O. Box 225916D.,1133, TX 75265
. Al len MunroBehavioral Technology Labor at,r1ea1915 El area Ave Fourth FloorRedondo Beach, CA 90277
Dr. Donald A NormanTmpt of Psych logy C-009Univ. of Califorria, San DiegoLa Jolla, CA 92091
Dr. Melvin R. Novick355 Lindqui at Center for Me saurrsentUniversity of IowaVsGa City. IA 52202
Dr. Jere Or lanskyIn''!.1tute for Defense Analyses
ray Navy Dr I v e":ntor,. VA 22202
1
t
'04) yntorr A. raps tn,asaehuletta I I It 14 lite of 1,artificial IntelllAvnr r tat.041.1 Terlan,Ithly `Alurst e
arbc1)1iie MA (1.'1
Pr . Jears A. Para attnP' t 1 sod to Unto', altP.D. Brit 7,0err t I )". PH
'AT
No) ta, v
P. I i'41'10141
c/o III y a 1). r 0,1))- thrrot1.1411ver alt al 1f7rnlaIre, kelp, CA 901.'0
. Andrew II, RoseAin,' lean [nal I! . tars for Res-a, h
10')", Thomas le Ter 5.)))) St. W41
Al 1111141to0 141 :0011!
1 MR. (DIG/ Pl. Tat),1() lilt 1. R )thlro f,'61t M. Et1lIW-4,D INIEF Bell LJF.,r stor leaAA). !V;T(''4, VA 2e.'0! )))1') ItatalTa In As ..))».
.'ay HI 1 I NJ ')'))A1 Dr. mar ths Poison
Petra, tareqt of Psy,ho 1 44yUnlversIty .fPOUti.r, Cr) A)) ID.'
1 DR, PF TIN (')1...DEP/ . 0E Pr.TC.I11
UNIVr RI!'" S.,-)Lla ,r)ROW,. '41, -0 Al,n1
Dr, I t rDe par trs ,
Univert)14, )( Nonve-Denvoc.C.1 .1)),.))))
DA. ea SET viR-K RESEARCH 1 )-7'm39,7 RID 11 1.4mAl c 0'
m1NR AT M.P I T 41
9UN'III.7.4 1 NPOD AC'rD-53 BONN
41
1 Dr, mark D. PEA), at tonal Pa ychol ort y Dept .
trMI I1 1Ia 1 1
MO 55:'11
1 Dr (ruin aarnItaerIL ,.1 Pry. 1)logy
1ln iv er 51 t y ,f k))^11illAtAm,rattle. HA 'IF' t'.
rttl TER !)'-.)14r-
t) Li nN rs-.,-,,,,Y( ) I ;911";
Or. Alan Cl 1,1 en re LIDe lair amen, of MA the)), t It,Harr II ton '',11egeCl int,n , NY 11171
camrsIttee an ,iir) I Live 13, 3,,ar
I Dr Lonnie R. Sherrotlla I Frc lent-, brie arc h Cole) 11
505 Third Avenue9r4 York, NY 10,715
Robe-t C. SieglerAsa,-late Profess."'Carne0 le- Mello no Iv er31) ty(Mpart.ment. of Pay.cl-srlogy2-ohenley Parkl'1 t shu-gh. PA 15213
Dr. rta^).1 E. 3sith4., It II r.,-0-k , Inc.4-)))) moul tot, CT) reel.CrtnbrIdge 110 12139
14.0) Ia., I
In II .1)ro I 'a,a, t r et Irma
RulAn , 'lo Corr y
11.1 o1901
. At, hat I ot.)1...)1,))1 of I.)), at inI.tanford 114) lye)) llt.yI.t an fo rd , CA ')1 105
. 14,1)er t Mter ober gt + -pt.. of PsychologyTale ,hllvrritty16,1 11A, Yale st a 1,r1New Haw en , CT Of)57
Irtuit I A 1 loan).Illly . of caltr,nt.1r0liss iorI ye) hoology18115 'I, II en' Ave.Notion h. 11,11,11, CA VI,' I
(4- J, one r1'p r' eparon Ica. Inc.:' / I Vat lel Anent...
u., 1 1 and 11111a CA 91 164
. 41111 arm R. III 1.1v er t y .)( Ilichlgor
',It It ute for nocial ao,r) hAnn A.- b or N I 611101,
1 DR. ALBERT STEVEIr. , Fhuard 4,11ner[VOLT BERANEK & NEVPIAN INC. isjre ,r SCirnce Or 1) a) II)4) MOULTON STREET 17911 II str,rt q, 4.
CAMBRIDGE, MA 02119 Ws an I ng tOn ,01,1It.
1 Dr. Thom n s S. St 1 cDirer for , Itl sic skull ill,131on(NRDO
',A)) N. ahing ton :-.treetAl et andr is , VA
1 Davl1 E ant', Ph .1.Ha re 1 t Inc C..or(nrat Ion'6141 Olaf r.pr inghooa, PoM,.:,. an VA 22102
' PATRICK SAPPED:ITIITE FOR HATHEMATICAt 'N
THE SOCIAL SCIENCE,nTANFORI, UNIVERSITY
/A141-"RD, CA 941105
1 Dr KI14)/0 I TatsurokaCornpu: et- Rased Educat for search
Lab'or a tor y252 Fr glr11,1)r Ins Research Labor ator yUnlve-ay_y of IllinoisUrhas , IL 51801
C ivl.1 Thl 35en7,-par Lment of Psychology-rivet-51[y of KAnsasL wrence. KS 650441
T CO _ ABLEBE S. J.
1', yl 1 a Weaver.1444 :-.),..t11,1 Of t Ion
7.arvar)1 On iver5lty;OD Larsen Hal 1 Appi an Way
Idge n,"179
Or, kav id J. 4131N650 Elliott Kt))OrliVer 31 Ty of75 E. River R.r.k11.41)-leap0,1 I a MN 551`,5
fir, Ke It') T. Wescou)Inforlaat 1,n fences Dept.The Ran-1 Corporation1700 Main StCanto Nunt,..1. CA q0105
DR. SII3AN ' WHITELYPSYC)IOLOGY 10 P44411/1E/ST
UNIVERSITY Is /4A5LAWRENCE. KA4SAt 5600)
. stopher WickensDe oar talent of Psych logylAilvectsity ofCIIharrpaign. /1. 618711
1 Or, 3, Arthur 4ocHwardDep.), talent of PsychologyUniv erslty of Californiajt
J 44,4 5,4 t 0,4. rc0yy